Algorithm for calculating the forecasted passenger flow of innovative railway transport based on a neural model

Aim: to create a neural network toolkit for predicting passenger traffic by innovative (high-speed) rail transport. Materials and methods: statistical data on passenger traffic on all modes of transport in the areas of high-speed railway lines. Methods: econometric modeling, multilayer perceptron. Results: an algorithm for planning passenger traffic on high-speed railway lines based on an intelligent neural model. Conclusion: The ability of the neural network model to "learn" allows to increase the reliability of population mobility forecasts when evaluating HSRT projects.